The role of model dynamics in ensemble Kalman filter performance for chaotic systems
The ensemble Kalman filter (EnKF) is susceptible to losing track of observations, or ‘diverging’, when applied to large chaotic systems such as atmospheric and ocean models. Past studies have demonstrated the adverse impact of sampling error during the filter’s update step. We examine how system dyn...
| Main Authors: | , , , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Co-Action Publishing
2014
|
| Online Access: | http://hdl.handle.net/1721.1/89042 https://orcid.org/0000-0002-8362-4761 |
| _version_ | 1826199277480706048 |
|---|---|
| author | NG, GENE-HUA CRYSTAL MCLAUGHLIN, DENNIS ENTEKHABI, DARA AHANIN, ADEL McLaughlin, Dennis Entekhabi, Dara Ahanin, Adel |
| author2 | Massachusetts Institute of Technology. Department of Civil and Environmental Engineering |
| author_facet | Massachusetts Institute of Technology. Department of Civil and Environmental Engineering NG, GENE-HUA CRYSTAL MCLAUGHLIN, DENNIS ENTEKHABI, DARA AHANIN, ADEL McLaughlin, Dennis Entekhabi, Dara Ahanin, Adel |
| author_sort | NG, GENE-HUA CRYSTAL |
| collection | MIT |
| description | The ensemble Kalman filter (EnKF) is susceptible to losing track of observations, or ‘diverging’, when applied to large chaotic systems such as atmospheric and ocean models. Past studies have demonstrated the adverse impact of sampling error during the filter’s update step. We examine how system dynamics affect EnKF performance, and whether the absence of certain dynamic features in the ensemble may lead to divergence. The EnKF is applied to a simple chaotic model, and ensembles are checked against singular vectors of the tangent linear model, corresponding to short-term growth and Lyapunov vectors, corresponding to long-term growth. Results show that the ensemble strongly aligns itself with the subspace spanned by unstable Lyapunov vectors. Furthermore, the filter avoids divergence only if the full linearized long-term unstable subspace is spanned. However, short-term dynamics also become important as nonlinearity in the system increases. Non-linear movement prevents errors in the long-term stable subspace from decaying indefinitely. If these errors then undergo linear intermittent growth, a small ensemble may fail to properly represent all important modes, causing filter divergence. A combination of long and short-term growth dynamics are thus critical to EnKF performance. These findings can help in developing practical robust filters based on model dynamics. |
| first_indexed | 2024-09-23T11:17:13Z |
| format | Article |
| id | mit-1721.1/89042 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T11:17:13Z |
| publishDate | 2014 |
| publisher | Co-Action Publishing |
| record_format | dspace |
| spelling | mit-1721.1/890422022-09-27T18:26:51Z The role of model dynamics in ensemble Kalman filter performance for chaotic systems NG, GENE-HUA CRYSTAL MCLAUGHLIN, DENNIS ENTEKHABI, DARA AHANIN, ADEL McLaughlin, Dennis Entekhabi, Dara Ahanin, Adel Massachusetts Institute of Technology. Department of Civil and Environmental Engineering NG, GENE-HUA CRYSTAL McLaughlin, Dennis Entekhabi, Dara Ahanin, Adel The ensemble Kalman filter (EnKF) is susceptible to losing track of observations, or ‘diverging’, when applied to large chaotic systems such as atmospheric and ocean models. Past studies have demonstrated the adverse impact of sampling error during the filter’s update step. We examine how system dynamics affect EnKF performance, and whether the absence of certain dynamic features in the ensemble may lead to divergence. The EnKF is applied to a simple chaotic model, and ensembles are checked against singular vectors of the tangent linear model, corresponding to short-term growth and Lyapunov vectors, corresponding to long-term growth. Results show that the ensemble strongly aligns itself with the subspace spanned by unstable Lyapunov vectors. Furthermore, the filter avoids divergence only if the full linearized long-term unstable subspace is spanned. However, short-term dynamics also become important as nonlinearity in the system increases. Non-linear movement prevents errors in the long-term stable subspace from decaying indefinitely. If these errors then undergo linear intermittent growth, a small ensemble may fail to properly represent all important modes, causing filter divergence. A combination of long and short-term growth dynamics are thus critical to EnKF performance. These findings can help in developing practical robust filters based on model dynamics. National Science Foundation (U.S.) (CMF Program Grant 0530851) National Science Foundation (U.S.) (DDAS Program Grant 0540259) National Science Foundation (U.S.) (ITR/AP Program Grant 0121182) 2014-08-25T19:16:20Z 2014-08-25T19:16:20Z 2011-10 Article http://purl.org/eprint/type/JournalArticle 02806495 1600-0870 http://hdl.handle.net/1721.1/89042 NG, GENE-HUA CRYSTAL, DENNIS MCLAUGHLIN, DARA ENTEKHABI, and ADEL AHANIN. “The Role of Model Dynamics in Ensemble Kalman Filter Performance for Chaotic Systems.” Tellus A 63, no. 5 (September 15, 2011): 958–977. https://orcid.org/0000-0002-8362-4761 en_US http://dx.doi.org/10.1111/j.1600-0870.2011.00539.x Tellus A Creative Commons Attribution http://creativecommons.org/licenses/by/4.0/ application/pdf Co-Action Publishing Co-Action Publishing |
| spellingShingle | NG, GENE-HUA CRYSTAL MCLAUGHLIN, DENNIS ENTEKHABI, DARA AHANIN, ADEL McLaughlin, Dennis Entekhabi, Dara Ahanin, Adel The role of model dynamics in ensemble Kalman filter performance for chaotic systems |
| title | The role of model dynamics in ensemble Kalman filter performance for chaotic systems |
| title_full | The role of model dynamics in ensemble Kalman filter performance for chaotic systems |
| title_fullStr | The role of model dynamics in ensemble Kalman filter performance for chaotic systems |
| title_full_unstemmed | The role of model dynamics in ensemble Kalman filter performance for chaotic systems |
| title_short | The role of model dynamics in ensemble Kalman filter performance for chaotic systems |
| title_sort | role of model dynamics in ensemble kalman filter performance for chaotic systems |
| url | http://hdl.handle.net/1721.1/89042 https://orcid.org/0000-0002-8362-4761 |
| work_keys_str_mv | AT nggenehuacrystal theroleofmodeldynamicsinensemblekalmanfilterperformanceforchaoticsystems AT mclaughlindennis theroleofmodeldynamicsinensemblekalmanfilterperformanceforchaoticsystems AT entekhabidara theroleofmodeldynamicsinensemblekalmanfilterperformanceforchaoticsystems AT ahaninadel theroleofmodeldynamicsinensemblekalmanfilterperformanceforchaoticsystems AT mclaughlindennis theroleofmodeldynamicsinensemblekalmanfilterperformanceforchaoticsystems AT entekhabidara theroleofmodeldynamicsinensemblekalmanfilterperformanceforchaoticsystems AT ahaninadel theroleofmodeldynamicsinensemblekalmanfilterperformanceforchaoticsystems AT nggenehuacrystal roleofmodeldynamicsinensemblekalmanfilterperformanceforchaoticsystems AT mclaughlindennis roleofmodeldynamicsinensemblekalmanfilterperformanceforchaoticsystems AT entekhabidara roleofmodeldynamicsinensemblekalmanfilterperformanceforchaoticsystems AT ahaninadel roleofmodeldynamicsinensemblekalmanfilterperformanceforchaoticsystems AT mclaughlindennis roleofmodeldynamicsinensemblekalmanfilterperformanceforchaoticsystems AT entekhabidara roleofmodeldynamicsinensemblekalmanfilterperformanceforchaoticsystems AT ahaninadel roleofmodeldynamicsinensemblekalmanfilterperformanceforchaoticsystems |